Max Robotics — Global Service Robot Directory + US Certification

About

Peter Decker’s research bridges computer vision and autonomous construction robotics, with a focus on robust motion estimation and multi-robot coordination. His early work on egomotion estimation tackled the critical challenge of degeneracy in essential matrix calculations for visual odometry—a foundational problem for augmented reality and robotic navigation. That 2008 paper, though with modest citation counts, addresses a subtle but persistent issue in 3-D pose estimation from single cameras, influencing subsequent work in the field. More recently, Decker has advanced autonomous road construction, co-authoring studies on distributed coordination and task assignment for tandem rollers (2019) and behavior-based edge compaction for off-road robots (2021). These contributions demonstrate how robotic skills can be extended to real-world, unstructured environments, enabling fleets of autonomous rollers to collaborate efficiently on construction sites. While his citation impact is still growing, Decker’s work is notable for its practical, application-driven approach—tackling both the theoretical underpinnings of visual estimation and the deployment of multi-agent systems in heavy civil engineering. His research offers valuable insights for students interested in the intersection of perception, control, and field robotics.